Analysis of monitoring data for the safety control of dams using neural networks
Concrete dam deformations are usually managed using a system of instruments set at various sections which control dam inclination, air and concrete temperature, water level and weather conditions. The study of monitored data is of the outmost importance for the assessment of both the structure and the basin. The implementation of mathematical model to correlate ambient variables and dam deformations is performed in order to gain a forecasting tool able to define critical thresholds for dam management. The present paper is aimed at presenting the artificial neural networks (ANN) as a powerful mathe- matical model used to foresee dam inclinations considering values of a set of ambient variables. The ANN model is applied to a case study, the dam of Pieve di Cadore, and comparison with a conventional statistical regression model is finally presented.
artificial neural networks dam monitoring data forecast
A.Panizzo A.Petaccia
DITS Department, La Sapienza University, Rome, Italy Italian National Dam Registry, Rome, Italy
国际会议
The Fifth International Conference on Fluid Mechanics(第五届国际流体力学会议)
上海
英文
2007-08-15（万方平台首次上网日期，不代表论文的发表时间）